s1ghhh/VLADrop_LingbotVLA_LIBERO_Baseline

VLADrop-LingbotVLA-LIBERO-baseline

Checkpoint for Drop-Then-Recovery: How Redundant Are Vision-Language-Action Models?.

DTR (Drop-Then-Recovery) removes transformer blocks from a pretrained VLA model and recovery-fine-tunes the smaller dense model. Code: https://github.com/s1ghhh/VLADrop

This checkpoint

Paper row	Table 7: Lingbot-VLA baseline (0/36 dropped)
Dropped blocks	none (full model)
Recovery training	batch size 16, 50K steps, lr 5e-5, fine-tuned from lingbot-vla-4b on standard LIBERO
LIBERO success rate	Spatial 81.8 / Object 95.0 / Goal 86.6 / Long 67.8 / Avg 82.8

Usage

Lingbot-VLA HF-format checkpoint. Use with the VLADrop lingbot-vla code (https://github.com/s1ghhh/VLADrop): serve with python -m deploy.lingbot_libero_policy --model_path <this_repo_local_path> --port 8200 --use_length 8 then run experiment/libero/libero/run_libero_eval.py --model_family instruct_vla. Requires the Qwen2.5-VL-3B-Instruct tokenizer (env QWEN25_PATH). The drop lists are recorded in the model config produced at export time.

Citation

@article{sun2026vladrop,
  title={Drop-Then-Recovery: How Redundant Are Vision-Language-Action Models?},
  author={Sun, Guoheng and Feng, Kaixi and He, Shwai and Gong, Xiaochuan and He, Yexiao and Wang, Ziyao and Shen, Zheyu and Ye, Wanghao and Kompella, Ramana Rao and Liu, Gaowen and Li, Ang},
  journal={arXiv preprint arXiv:2606.27755},
  year={2026}
}